International Journal of Wireless & Mobile Networks (IJWMN) Vol. 7, No. 4, August 2015 DOI : 10.5121/ijwmn.2015.7405 63 WINDOW BASED SMART ANTENNA DESIGN FOR MOBILE AD HOC NETWORK ROUTING PROTOCOL AKM Arifuzzman 1 , Rumana Islam 2 , and Mohammed Tarique 3 1 Department of Electrical Engineering, University of Alabama, Birmingham, AL, USA 2 Department of Electrical and Electronic Engineering, American International University, Banani, Dhaka,Bangladesh 3 Department of Electrical Engineering, Ajman University of Science and Technology, Fujairah, United Arab Emirates ABSTRACT Mobile Ad hoc Networks (MANETs) have drawn considerable attentions of the researchers for the last few years. MANETs, consisting of mobile nodes, are self-organizing and self-configuring and hence can be deployed without any infrastructure support. MANETs also have some limitations including short-life, unreliability, scalability, latency, high interference, and limited resources. In order to overcome these limitations many innovations and researches have been done in this field. Incorporating smart antenna system with the mobile nodes is one of them. It has been shown in the literatures that smart antenna can improve network capacity, increase network lifetime, reduce delay, and improve scalability by using directional radiation pattern. But, there are some unsolved issues too. Smart antenna requires a large number of antenna elements that a resource constraint mobile node can hardly handle. Hence, one major design issue is to achieve a desired radiation pattern by using minimum number of antenna elements. Another important issue is the arrangement of antenna elements. Antenna elements can be arranged in linear, planar, and circular manners. In this paper we have addressed these issues. We have proposed a window based smart antenna design for MANETs. Our target is to improve the routing performance of MANETs. We have shown that by using appropriate window function a desired radiation pattern can be achieved with a minimum number of antenna elements. KEYWORDS Ad hoc networks, routing protocol, DSR, smart antenna, radiation, window, energy constraint, antenna array 1. INTRODUCTION Wireless networking has been an active field of research since the early days of packet radio network introduced by Defense Advanced Research Project Agency (DARPA) [1]. Recent developments in wireless devices and applications have caused exponential growth in wireless users. Wireless devices such as laptop computers, personal digital assistant (PDA), pagers, and cellular phones have become portable now. Users can carry these devices anywhere at any time. Hence, there is a need for a network that can be readily deployed at any place at any time without any centralized administration. In many scenarios infrastructure based network is hard to build and maintain. Wireless network operating in a battlefield (see Figure 1) is an example for such scenarios. In this scenario an infrastructure based network cannot support communication among
18
Embed
WINDOW BASED SMART ANTENNA DESIGN FOR ...airccse.org/journal/jwmn/7415ijwmn05.pdf2Department of Electrical and Electronic Engineering, American International University, Banani, Dhaka,Bangladesh
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 7, No. 4, August 2015
DOI : 10.5121/ijwmn.2015.7405 63
WINDOW BASED SMART ANTENNA DESIGN
FOR MOBILE AD HOC NETWORK ROUTING
PROTOCOL
AKM Arifuzzman1, Rumana Islam
2, and Mohammed Tarique
3
1Department of Electrical Engineering, University of Alabama, Birmingham, AL, USA
2Department of Electrical and Electronic Engineering, American International University,
Banani, Dhaka,Bangladesh 3Department of Electrical Engineering, Ajman University of Science and Technology,
Fujairah, United Arab Emirates
ABSTRACT
Mobile Ad hoc Networks (MANETs) have drawn considerable attentions of the researchers for the last few
years. MANETs, consisting of mobile nodes, are self-organizing and self-configuring and hence can be
deployed without any infrastructure support. MANETs also have some limitations including short-life,
unreliability, scalability, latency, high interference, and limited resources. In order to overcome these
limitations many innovations and researches have been done in this field. Incorporating smart antenna
system with the mobile nodes is one of them. It has been shown in the literatures that smart antenna can
improve network capacity, increase network lifetime, reduce delay, and improve scalability by using
directional radiation pattern. But, there are some unsolved issues too. Smart antenna requires a large
number of antenna elements that a resource constraint mobile node can hardly handle. Hence, one major
design issue is to achieve a desired radiation pattern by using minimum number of antenna elements.
Another important issue is the arrangement of antenna elements. Antenna elements can be arranged in
linear, planar, and circular manners. In this paper we have addressed these issues. We have proposed a
window based smart antenna design for MANETs. Our target is to improve the routing performance of
MANETs. We have shown that by using appropriate window function a desired radiation pattern can be
achieved with a minimum number of antenna elements.
KEYWORDS
Ad hoc networks, routing protocol, DSR, smart antenna, radiation, window, energy constraint, antenna
array
1. INTRODUCTION
Wireless networking has been an active field of research since the early days of packet radio
network introduced by Defense Advanced Research Project Agency (DARPA) [1]. Recent
developments in wireless devices and applications have caused exponential growth in wireless
users. Wireless devices such as laptop computers, personal digital assistant (PDA), pagers, and
cellular phones have become portable now. Users can carry these devices anywhere at any time.
Hence, there is a need for a network that can be readily deployed at any place at any time without
any centralized administration. In many scenarios infrastructure based network is hard to build
and maintain. Wireless network operating in a battlefield (see Figure 1) is an example for such
scenarios. In this scenario an infrastructure based network cannot support communication among
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 7, No. 4, August 2015
64
soldiers, warships, fighter planes, tanks, and other combating equipment because these
constituents of the network are moving at different directions in an unpredictable manner. There
are also some cases where infrastructure may not exist to build a network on it due to natural
calamity like cyclone, tsunami, and tornado. Hence, there is always a need to set up and maintain
a temporary network among a group of users without any pre-existing infrastructure. Mobile Ad
hoc Network (MANET) is considered a suitable solution for this kind of temporary network.
MANET is consisting of a group of mobile nodes, which have limited battery capacity and have
limited processing power. MANET is self-organizing and self-configuring. Initially, MANET has
been developed to provide networking support in military applications, where infrastructure based
network is almost impossible to set up and maintain. MANET also has other numerous
applications such as crisis management, telemedicine, tele-geoprocessing, process control,
personal communication, virtual navigation, education, and security [2]. These applications
impose diversified design and performance constraints on MANET. In contrast to its wired
counterpart MANET has many unique characteristics. It has dynamic topology. Mobile nodes can
join and leave the network at any time. Hence route ‘breakage’ is a very frequent phenomenon in
MANET. Since the medium of communication in MANET is wireless, it has high packet loss,
inherent unreliability, high interference, and noise. Two other most addressed problems are
limited capacity and short transmission range. MANET has limited bandwidth because mobile
nodes share the wireless medium. Mobile nodes communicate with each other in a multi-hop
fashion due their limited transmission range.
Figure 1 Mobile Ad hoc Networks (MANET)
Numerous research activities have been conducted in order to improve the performances of
MANETs. Smart antenna integration with mobile node is one of them. In most of the applications
mobile nodes in a MANET are equipped with omni-directional antennas. With omni-directional
antenna a mobile node radiates power equally in a surrounding area and all mobile nodes within
the transmission range receive that signal. Due to the omni-directional radiation and imposed
medium access control (MAC) algorithm the neighboring nodes remain standby during their
transmissions to avoid packet collision. In order to overcome this limitation there has been a
rapidly growing interest in the use of smart antenna system in MANET [3-12]. Smart antenna
system has the ability to radiate signal in a given direction. By using smart antenna a mobile node
can adjust its transmission power to a required minimum level and hence can improve the
network life-time. Smart antenna also reduces interference level in a network by steering antenna
radiation’s nulls toward the sources of interference and can support a number of simultaneous
transmissions as shown in Figure 2.
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 7, No. 4, August 2015
65
There are also other advantages of using smart antenna in MANET. These advantages are
addressed in [16]. In this paper we limit our effort only to investigate the effects of smart antenna
on the routing protocol of MANET. There are many routing protocols have been proposed for
MANET. Among these protocols Dynamic Source Routing (DSR) protocol is a popular one [17].
In this paper we propose a smart antenna system for MANET to improve the performance of the
DSR routing protocol. The rest of the paper is organized as follows. Section 2 presents a brief
introduction to smart antenna. Section 3 presents the basic operation of DSR protocol. The effects
of smart antenna on the routing protocol have been described in section 4. Section 5 contains the
smart antenna design issues for MANET. Section 6 introduces window functions and section 7
describes the side lobe minimization techniques by using window functions. The design trade-offs
have been discussed in section 8. This paper is concluded with section 9.
Figure 2 Capacity improvement of ad hoc network with smart antenna [10]
2. SMART ANTENNA
Smart antennas are composed of a collection of two or more antenna elements called antenna
array. These antenna elements work together to establish a unique radiation pattern in a desired
direction. A functional block diagram of a smart antenna system is shown in Figure 3. It is
depicted in the figure that digital signal processor is the backbone of such system. Smart antenna
system can locate and track desired signal called Signal of Interest (SOI) and also Signal of no
Interest (SNOI). The smart antenna system can dynamically adjust the radiation of the antenna
elements so that the resultant radiation is directed along the SOI and it can suppress the radiation
pattern along the direction of SNOIs. In order to track the radiation pattern direction of arrival
(DOA) or angle of arrival (AOA) algorithms is used. The DOA algorithm helps the antenna
elements to set the desired excitation weights in terms of magnitude and phase so that the
radiation can be directed in a desired direction.
Figure 3 Smart antenna system
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 7, No. 4, August 2015
66
Antenna array can be arranged in many different ways. Most popular arrangements are linear
array, circular array, and planer arrays. These arrangements are associated with different level of
complexity in terms of hardware and software. Since mobile nodes in ad hoc networks are
constrained with limited processing power and limited battery life, it is imperative to choose a
suitable arrangement. In this paper we focus on two main issues. First, we have investigated
different types of antenna arrangement for the mobile nodes so that the required radiation pattern
can be achieved with minimum complexity. Second, we have investigated the required number of
antenna elements for each type of arrangement. With too few antennas desired radiation pattern
cannot be achieved. On the other hand too many antennas should not be selected for resource
constraint mobile nodes. In this paper we investigate these issues so that an optimum number of
antennas can be arranged in a proper arrangement and the performance of DSR routing protocol
can be improved.
3. DYNAMIC SOURCE ROUTING PROTOCOL
The DSR protocol is an on-demand reactive routing protocol [17]. It consists of two main
mechanisms namely (a) route discovery, and (b) route maintenance.
Route discovery is a mechanism by which a source node finds a route to a destination. If a source
node wants to send some packets to a destination, it first searches its route cache to find a route. If
the source node cannot find a route there, it initiates the route discovery process by transmitting a
route request message. Each route request contains unique route request ID in addition to the
source and destination addresses. All other nodes, within the range of the source, receive the
request message. These nodes check the destination address of the request message. If the
destination address does not match with the node’s own address, it appends its address in the
route request and re-broadcast the message. This process goes on unless the route request reaches
its intended destination. Once the desired destination node receives the route request, it sends a
reply packet to the source. The destination node copies the accumulated routing information from
the route request packet to the route reply packet. When the source node receives the route reply
packet, it records the newly discovered route in its route cache and starts sending data packet via
this route. A typical route discovery mechanism is shown in the Figure 4. Here, the source node S
initiates a route discovery to discover the route to the destination D. Mobile nodes A and E
receive the route request message. Since these two nodes are not the destination, they add their
addresses in the request packet and forward that request packets to their neighbors. This process
goes on until the request packet is finally received by the destination node D, which sends reply
to the source. In the scenario two routes have been discovered namely S-A-B-C-D and A-E-F-G-
H-I-D. While generating the route reply packet, the destination copies these two routes from the
route request packet to route reply packet. Once the reply packets are received by the source, it
records these two discovered routes in the route cache and starts sending packet by using the
shorter discovered route (i.e., S-A-B-C-D).
Figure 4 Route discovery of DSR protocol
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 7, No. 4, August 2015
67
Route maintenance mechanism helps a mobile node to detect changes in the network topology. If
a node cannot send a packet to the destination due to ‘broken’ link, it initiates the route
maintenance mechanism. By using this mechanism a mobile node sends a message to the source
node (and other nodes on the same route) about the ‘broken’ link. The DSR protocol uses
acknowledgment mechanism of the underlying Medium Access Control (MAC) protocol for the
‘broken’ link detection. For example, IEEE 802.11 Wireless LAN medium access layer provides
acknowledgement for each packet. If the transmitting node does not receive any
acknowledgement after sending a packet several times through a link, it treats that link as
‘broken’ and updates its route cache by marking the route as ‘invalid’. A typical route
maintenance mechanism is illustrated in Figure 5. In this scenario it is assumed that mobile C
exhausts battery and hence the link B-C is ‘broken’. Node B tries to send packets to C for several
times and node C sends no acknowledgement packet. The node B assumes that the link B-C is
broken. In order to let the source S and other nodes (i.e., B and A) lying on the same route, mobile
node B generates a route error message and sends it to these nodes. Once route error message is
received by the source S, it marks the route S-A-B-C-D invalid in the route cache and starts
sending data packets by using the other alternative route (i.e., A-E-F-G-H-I-D).
Figure 5 Route maintenance mechanism of DSR protocol
4. DSR PROTOCOL IMPROVEMENT BY SMART ANTENNA
The route request mechanism of DSR routing protocol can be made more efficient by using smart
antenna. In DSR, a node can learn and cache multiple routes to a destination by means of a single
route discovery. To ensure this multiple routing strategy work all neighboring nodes are obligated
to re-broadcast once they receive a route request. The ultimate outcome of the re-broadcasting is
’flooding’ of overhead packets in the network. Although some measures have been adopted in the
DSR protocol to reduce flooding such as limiting the rate of route discovery by using random
back-off algorithm, and imposing shorter hop count (’ring zero search’ mechanism); flooding
problem is still severe in the DSR protocol specially for a large network. Some of the drawbacks
related to flooding are redundant re-broadcast, contention, and collision.
Figure 6 Broadcast reduction
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 7, No. 4, August 2015
68
The redundant broadcast occurs when a mobile node re-broadcasts a route request message to its
neighbors, which may have already received that message from other nodes. The contention
problem occurs when all neighbors of a node re-broadcast the request packet at the same time.
These neighbors may severely contend with each other to get access to the medium. The
collisions are more likely to occur when all neighboring nodes try to re-broadcast at the same
time. Such kind of broadcasting is illustrated in Figure 6. In this scenario node A is the source and
node D is the destination. When node A initiates the route discovery mechanism by broadcasting
the route request packet, all the neighboring nodes namely B, C, and G receive the request packet
and re-broadcast. Hence, there will be six copies of the request messages in the network. This
type of redundant message can be easily reduced by using smart antenna. Let us assume that node
G, equipped with smart antenna, can determine the angle of arrival (AOA) of the request message
once it receives from the source node A. Hence, node G should broadcast in a direction that is in
opposite of arrival direction estimated from the received request message. Finally, the destination
node receives the request packet broadcast by the node G and hence sends reply back to the
source A. The reply packet carries the information about the route A-G-D.
Smart antenna can also help to efficiently handle route reply packet. For example when the node
D generates a route reply packet, it sends the packet through the route D-G-A. Since the node D
and G already know the direction of the arrivals (DOAs) from the route requests, these nodes
should send the reply packet in a desired direction. Hence, the route reply packets are not going to
affect the other nodes (i.e., E,F,B, and C) transmissions; and these nodes can perform their normal
network operation.
Figure 7 Data packet delivery
Once a route is discovered a source node starts sending data packet by using the discovered route.
This kind of data packet delivery is illustrated in Figure 7. In this scenario node A is the source of
the data packet. Node C is the destination node and node B is a forwarding node (i.e., router).
Since the AOAs are known from the route reply packet, node A adjusts its antenna elements’
weights accordingly. Hence, the resultant radiation pattern’s main lobe will be directed toward
node B. The node B also adjusts the weights of the antenna elements so that it can direct the
radiation pattern toward node C. Since all the nodes involved in the data transmission have
appropriately adjusted the antennas weights, their transmissions will not affects other surrounding
nodes.
The smart antenna also helps in handling the route error message. Assume that the link between
node B and C is ‘broken’ for some reason. The node B generates a route error message and sends
it to node A to let it know that the route B-C is broken. While sending the route error message
node B radiates the signal in the desired direction so that other nodes are not affected.
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 7, No. 4, August 2015
69
Although smart antenna system helps the MANETs to improve the routing performance as
discussed above, it has some drawbacks too. One of the drawbacks is that mobile nodes should be
equipped with transceivers that are much more complex than traditional transceiver. Determining
the necessary weights for antenna elements is also computationally extensive and hence a very
powerful processor needs to be associated with the mobile nodes. Moreover, mobile nodes should
minimize the side lobes. Although the smart antenna helps the mobile nodes to direct main
radiation lobe in a desired direction, there is always a chance of generating additional minor lobes
as shown in Figure 7. One of the solutions for minimizing the minor lobes is to increase the
number of antenna elements. This may not always be feasible because of the high cost and
complexity associated with the increasing number of antenna elements. Hence, the main design
objective will be to have a desired radiation pattern with the minimum number of antenna
elements. Moreover, the antenna arrangement should be as simple as possible so that it can be
easily equipped with the nodes.
5. SMART ANTENNA DESIGN FOR MANETS
In smart antenna system the antenna elements may be arranged in many different ways. Most
common arrangements are linear, rectangular, and circular. Among these antenna array
arrangements linear array is the simplest one. In this arrangement all antenna array elements are
aligned along a straight line as shown in Figure 8. For the simplicity of our analysis we assume
that the antenna array is composed of isotropic radiating antenna elements and the antenna
elements are equally spaced by a distance d. We also assume that the far-field condition is
satisfied (i.e., r ˃˃ d). In general the array factor is given by